Elsevier

Journal of Biomechanics

Volume 48, Issue 7, 1 May 2015, Pages 1300-1309
Journal of Biomechanics

Hemodynamic transition driven by stent porosity in sidewall aneurysms

https://doi.org/10.1016/j.jbiomech.2015.02.020Get rights and content

Abstract

The healing process of intracranial aneurysms (IAs) treated with flow diverter stents (FDSs) depends on the IA flow modifications and on the epithelization process over the neck. In sidewall IA models with straight parent artery, two main hemodynamic regimes with different flow patterns and IA flow magnitude were broadly observed for unstented and high porosity stented IA on one side, and low porosity stented IA on the other side. The hemodynamic transition between these two regimes is potentially involved in thrombosis formation. In the present study, CFD simulations and multi-time lag (MTL) particle imaging velocimetry (PIV) measurements were combined to investigate the physical nature of this transition. Measurable velocity fields and non-measurable shear stress and pressure fields were assessed experimentally and numerically in the aneurysm volume in the presence of stents with various porosities. The two main regimes observed in both PIV and CFD showed typical flow features of shear and pressure driven regimes. In particular, the waveform of the averaged IA velocities was matching both the shear stress waveform at IA neck or the pressure gradient waveform in parent artery. Moreover, the transition between the two regimes was controlled by stent porosity: a decrease of stent porosity leads to an increase (decrease) of pressure differential (shear stress) through IA neck. Finally, a good PIV–CFD agreement was found except in transitional regimes and low motion eddies due to small mismatch of PIV–CFD running conditions.

Introduction

Flow diverter stents (FDSs) are endovascular treatments (Byrne et al., 2010, Pereira et al., 2014b) of sidewall intracranial aneurysms (IAs) (Brisman et al., 2006). These expandable tubular devices made of dense mesh of filaments (Fig. 2) are implanted in the parent artery to cover the IA neck, hence reducing the intra-aneurysmal flow known to promote IA thrombogenesis (Zanaty et al., 2014b, Pereira et al., 2014a). Despite the high success rate of FDS treatments, some complications partly related to hemodynamics issues, e.g. subacute aneurysm rupture, parent artery occlusion or flow persistency, have been reported (Brinjikji et al., 2013, Zanaty et al., 2014a). Recently, hemodynamic indicators measured per-operatively with digital subtracted angiography (DSA) (Pereira et al., 2013a, Chien and Vinuela, 2013) or assessed numerically with computational fluid dynamics (CFD) (Pereira et al., 2014c, Mut et al., 2015, Kulcsar et al., 2012, Zhang et al., 2013) were correlated with successful treatment outcomes. These results highlighted the importance of IA flow modifications in the healing process.

During the last decade, the effect of stent implantation in idealized sidewall IA models with straight parent artery was broadly investigated experimentally (Augsburger et al., 2009, Liou and Li, 2008, Liou et al., 2008, Lieber et al., 2002, Yu and Zhao, 1999, Yu et al., 2012, Bouillot et al., 2014a, Bouillot et al., 2015) and numerically (Hirabayashi et al., 2006, Hirabayashi et al., 2004, Kim et al., 2010, Appanaboyina et al., 2008, Bouillot et al., 2015). Although these studies considered various stent configurations within IAs with different size and shape, identical hemodynamic features were reported regardless of the inflow conditions.

  • For unstented and high porosity stented IA, an inflow jet at the distal part of the IA neck driving a large vortex was typically observed in IA.

  • When decreasing the stent porosity, the IA flow pattern was strongly altered. In general, a diffuse flow was observed at the IA neck with proximal (distal) inflow (outflow) inverted compared with unstented IA.

The transition between these two hemodynamic regimes was controlled by the stent porosity and led to a strong flow reduction potentially involved in thrombosis formation. Indeed, in patient specific IA geometries a similar transition has been predicted with CFD when decreasing the porosity of virtual stent (Shobayashi et al., 2013) and in successfully treated IA with FDS (Zhang et al., 2013). A detailed description of the underlying physics would be therefore very useful for the understanding of thrombosis initiation and design of new devices.

Recently, we proposed in Bouillot et al. (2014a) a possible physical mechanism describing this hemodynamic transition. Our hypothesis schematized in Fig. 1 involved both shear stress and pressure differential at the IA neck difficult to access experimentally. In the present study, we aim to show the evidence of this description supported with quantitative arguments and hence highlight the universal character of the transition. For this purpose, our recent particle imaging velocimetry (PIV) investigations (Bouillot et al., 2014a) were complemented with CFD simulations providing the non-measurable shear stress and pressure fields. These quantities were analyzed in both unstented/stented idealized IA with different porosities and correlated with IA flows measured by PIV and predicted by CFD.

Section snippets

Experimental and virtual IA models

Fig. 2 shows the idealized sidewall IA geometry considered in the CFD simulations and for the molding of the silicone phantom used in the PIV measurements. In such a model, the stent is expected to expand regularly minimizing the geometry uncertainties in both PIV measurement and CFD simulations. It is composed of a spherical aneurysm of radius R=5 mm located at distance d3=6mm below a straight cylindrical artery of radius r=2 mm. The inlet/outlet length d1=150mm/d2=110mm was set in the silicone

Hemodynamic description: PIV–CFD comparison

An overview of the measured and computed velocities for unstentend/stented IA with (S1–3) is provided in Figs. 3 (systole) and 4 (diastole) showing a comparison of the in-plane flow patterns and velocity magnitude as well as a 3D representation of the computed streamlines along with Q-isosurfaces highlighting the vortical structures. A movie of their time evolution during the full cardiac cycle is also available online. The two distinct regimes described in the introduction were observed:

  • In

Hemodynamic transition

We have identified various qualitative and quantitative evidences of the two shear and pressure mechanisms (Fig. 1) driving the two main hemodynamic regimes for unstented/high porosity and low porosity stented IA:

  • The flow patterns were in agreement with the expected IA flow in both regimes. In particular the inflow–outflow inversion at the neck with low porosity stent was located at the inversion point of pressure differential through the stent.

  • The waveform of SAVs followed the time dependence

Conclusion

By combining experimental and computational methods, we showed qualitative and quantitative evidences of the shear and pressure mechanisms driving the flow in unstented/high porosity and low porosity stented IA, respectively. These results suggest the universal behavior of the hemodynamic transition occurring in sidewall IA while decreasing the stent porosity and hence marks an important milestone in the understanding of IA hemodynamics.

Conflict of interest statement

None declared.

Acknowledgment

This study was supported by Swiss National Science Foundation grant (SNF 32003B_141192).

References (49)

  • L. Augsburger et al.

    Effect of flow diverter porosity on intraaneurysmal blood flow

    Clin. Neuroradiol.

    (2009)
  • L. Augsburger et al.

    Intracranial stents being modeled as a porous mediumflow simulation in stented cerebral aneurysms

    Ann. Biomed. Eng.

    (2011)
  • T. Aurboonyawat et al.

    An in vitro study of silk stent morphology

    Neuroradiology

    (2011)
  • A. Bernardini et al.

    Influence of different computational approaches for stent deployment on cerebral aneurysm haemodynamics

    Interface Focus

    (2011)
  • P. Bouillot et al.

    Particle imaging velocimetry evaluation of intracranial stents in sidewall aneurysmhemodynamic transition related to the stent design

    PLoS One

    (2014)
  • P. Bouillot et al.

    Multi-time-lag piv analysis of steady and pulsatile flows in a sidewall aneurysm

    Exp. Fluids

    (2014)
  • Bouillot, P., Brina, O., Ouared, R., Yilmaz, H., Lovblad, K., Farhat, M., Pereira, V., 2015. Computational fluid...
  • W. Brinjikji et al.

    Endovascular treatment of intracranial aneurysms with flow diverters a meta-analysis

    Stroke

    (2013)
  • J.L. Brisman et al.

    Medical progresscerebral aneurysms

    N. Engl. J. Med.

    (2006)
  • J.V. Byrne et al.

    Early experience in the treatment of intra-cranial aneurysms by endovascular flow diversiona multicentre prospective study

    PLoS One

    (2010)
  • A. Chien et al.

    Is flowmap, a novel tool to examine blood flow changes induced by flow diverter stent treatmentinitial experiences with pipeline cases

    J. Neurointerv. Surg.

    (2013)
  • X. He et al.

    Unsteady entrance flow development in a straight tube

    J. Biomech. Eng.—ASME

    (1994)
  • Y. Hoi et al.

    Validation of cfd simulations of cerebral aneurysms with implication of geometric variations

    J. Biomech. Eng.—ASME

    (2006)
  • C. Huang et al.

    Non-newtonian effect on hemodynamic characteristics of blood flow in stented cerebral aneurysm

    Commun. Comput. Phys.

    (2013)
  • Cited by (29)

    • Magnetic resonance imaging based modeling of microvascular perfusion in patients with peripheral artery disease

      2019, Journal of Biomechanics
      Citation Excerpt :

      However, the equations remain expressed in velocities rather than in changes of arbitrary signal intensities from CE-MRI scans. Several studies have previously reported on the use of a Newtonian fluid approximation for similar types of models (Javadzadegan et al., 2019, Vankan et al., 1998, Bouillot et al., 2015, Zhao et al., 2018, Bonfiglio et al., 2010, Boyd et al., 2007, Zhang et al., 2015). However, we have also performed the same models with a non-Newtonian fluid for 10 out of the 20 cases (5 PAD patients and 5 controls) in order to compare with the Newtonian fluid approach.

    • Four-Dimensional Phase Contrast Magnetic Resonance Imaging Protocol Optimization Using Patient-Specific 3-Dimensional Printed Replicas for In Vivo Imaging Before and After Flow Diverter Placement

      2017, World Neurosurgery
      Citation Excerpt :

      Understanding potential implications of flow patterns in cerebral aneurysms for disease progression and treatment efficacy is subject to ongoing research.1-10 Using computational fluid dynamics (CFD) techniques, a variety of approaches have been presented and selected hemodynamic parameters have been identified as potential candidates to help understand aneurysm rupture status and treatment efficacy after embolization with a flow diverter device (FDD).3,11-22 A matter of debate in this approach has been the accuracy of the computational models both regarding morphology and physiologic flow conditions, including segmentation accuracy and inflow and outflow boundary conditions.16,23-27

    • Near-wall hemodynamic parameters quantification in in vitro intracranial aneurysms with 7 T PC-MRI

      2023, Magnetic Resonance Materials in Physics, Biology and Medicine
    View all citing articles on Scopus
    View full text